PROJECT SUMMARY/ABSTRACT Mutations that activate phosphatidylinositol 3-kinase (PI3K) pathway signaling are among the most common events in human cancer. This has spurred tremendous therapeutic development seeking approaches to rationally inhibit this pathway in diverse malignancies. Despite these activities, and notwithstanding the recent dramatic increase in our knowledge of the spectrum, type, and frequency of PI3K pathway alterations in human cancer, broadly active targeted therapies for which biomarkers of sensitivity are available to facilitate patient selection are still lacking. We must address this urgent clinical challenge to improve the survival of cancer patients. We recently combined a unique institution-wide prospective clinical sequencing initiative with an active early-phase clinical trial program to demonstrate that an oral and selective ATP-competitive pan-AKT kinase inhibitor had significant single-agent activity in a multi-histology basket study of patients with AKT1 E17K-mutant advanced solid tumors. We went on to show in preliminary correlative studies that specific genomic correlates of response and resistance to AKT inhibition in AKT-mutant cancers are varied. These findings underscore the potential for developing unique therapeutic strategies targeting tumors with distinct mutational drivers that converge on hyperactivated PI3K signaling. Yet, without an integrated clinical- translational genomic approach to reveal the mechanisms underlying the dependence of solid tumors on mutant AKT, the gap in our understanding will only widen. We propose to overcome this challenge by establishing what conditions AKT inhibitor response and resistance in pre- and post-treatment tumor specimens and longitudinally collected tumor-derived cell-free DNA from patients with AKT-mutant human cancers. We will establish the genomic determinants of AKT inhibitor sensitivity. Then, we will identify which genomic abnormalities, when acquired or selected for, mediate AKT inhibitor resistance. Finally, we will identify and both functionally and clinically validate novel low-incidence activating mutations in AKT isoforms that confer similar pathway dependence and inhibitor sensitivity, thereby expanding a predictive biomarker for patient selection. Together, these studies seek to establish mutant AKT as a rational therapeutic target, and in doing so, establish a clinical-translational genomic framework to facilitate effective, evidence-based precision oncology in AKT-mutant cancers that can be extended to other molecularly defined populations of cancer patients.